This research aims to lower the drug use of 9- to 18-year-old American Indians. Process and outcome goals will be evaluated with representative samples of Native American counselors and adolescents. After a pool of Indian counselors has been recruited, a pilot sample of ten counselors will be pretested, trained, and posttested. Training curriculum will be revised, and 80 counselors will be randomly selected, pretested, matched on education and experience, and randomly assigned to training or to no-training control conditions. Training condition counselors will learn the prevention and early intervention strategy, and all counselors will be posttested. Indian adolescents will be recruited from schools, reservations, and Indian communities. A pilot group of ten Indian adolescents will be drawn, pretested, and treated by two trained counselors. When pilot data have upgraded prevention and early intervention curriculum, full-scale evaluation with 1,120 Indian adolescents will be launched. In four phases, samples of consenting adolescents will be randomly drawn, pretested, matched on age, and randomly assigned to intervention or no-intervention control conditions. After intervention groups of ten youths and two trained counselors have met for 24, 2-hour semiweekly sessions, all youths will be posttested. Process measures of counselor training and adolescent intervention will span demographic characteristics, the delivery of training and intervention, counselors' and youths' participation, and group member feedback. Pretest, posttest, and follow-up measures will quantify Indian counselors' and Indian adolescents' drug knowledge, attitudes towards drugs, locus of control, problem solving, coping, in vivo behavior, and assertiveness. Counselors will report their efforts to treat drug abuse. Adolescents will self-rate their personal esteem and future efficacy and report their drug avoidance, involvement with social networks, and substance abuse.